EXAR CORP - 10-K Annual Report

UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

WASHINGTON, D.C. 20549

FORM 10-K

ANNUAL REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934

For the fiscal year ended March 31, 2002

TRANSITION REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934 FOR THE TRANSITION PERIOD FROM ________ TO ________

Commission File No. 0-14225

EXAR CORPORATION

(Exact Name of registrant as specified in its charter)

Delaware		94-1741481
(State or other jurisdiction of incorporation or organization)		( I.R.S. Employer Identification No.)

48720 Kato Road, Fremont, CA 94538

(Address of principal executive offices, Zip Code)

Registrant’s telephone number, including area code: (510) 668-7000

SECURITIES REGISTERED PURSUANT TO SECTION 12(b) OF THE ACT: NONE

SECURITIES REGISTERED PURSUANT TO SECTION 12(g) OF THE ACT:

COMMON STOCK

(Title of Class)

Indicate by check mark whether the Registrant (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the Registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days. Yes x No ¨

Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K is not contained herein, and will not be contained, to the best of Registrant’s knowledge, in any definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K. x

The aggregate market value of the voting stock held by non-affiliates of the Registrant as of May 31, 2002 was $856,120,612 based on the last sales price reported for such date as reported on The Nasdaq Stock Market, Inc.

The number of shares outstanding of the Registrant’s Common Stock was 39,690,339 as of May 31, 2002, net of treasury shares.

DOCUMENTS INCORPORATED BY REFERENCE

Portions of the Company’s Definitive Proxy Statement to be filed not later than 120 days after the close of the fiscal year are incorporated by reference into Part III, Items 10,11,12 and 13 of this Report.

PART I

This Annual Report on Form 10-K (the “Annual Report”) contains certain forward-looking statements within the meaning of section 27A of the Securities Act of 1933, as amended, and section 21E of the Securities Exchange Act of 1934, as amended. These statements relate to, among other things, the Company’s future financial position, products, business development, strategy and management’s plans and objectives for future operations. Forward-looking statements generally can be identified by the use of forward-looking terminology such as “believe,” “may,” “will,” “intend,” “expect,” “estimate,” “continue,” “ongoing,” and “potential” or similar expressions or the negative of those terms or expressions. These statements involve known and unknown risks, uncertainties and other factors that may cause the Company’s actual results, performance or achievements to differ materially from those expressed or implied by such forward-looking statements. Readers are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date hereof. Factors that could cause actual results to differ materially from those included herein include, but are not limited to, the information contained under the captions “Part I, Item 1. Business,” and “Part II, Item 7. Management’s Discussion and Analysis of Financial Condition and Results of Operations,” and, in particular, “Risk Factors.” The Company disclaims any obligation to update information in any forward-looking statement.

ITEM 1. BUSINESS

Overview

Exar Corporation (“Exar” or the “Company”) designs, develops and markets high-performance, analog and mixed-signal silicon solutions for the worldwide communications infrastructure. Leveraging its industry-proven analog design expertise, system-level knowledge and standard process technologies, Exar provides OEMs with innovative, highly integrated ICs that facilitate the transport and aggregation of signals in access, metro and wide area networks. The Company’s physical layer silicon solutions address transmission standards such as T/E carrier, ATM and SONET/SDH. Additionally, Exar offers ICs for both the serial communications and the video and imaging markets. Exar’s customers include Alcatel, Cisco, Hewlett-Packard, Lucent, Nokia and Tellabs. Exar’s Common Stock trades on The Nasdaq Stock Market, Inc. under the symbol “EXAR” and is included in the S&P 600 SmallCap Index.

Industry Background

Communications technology has evolved from simple analog voice signals transmitted over networks of copper telephone lines to complex analog and digital voice and data signals transmitted over hybrid networks of media, such as copper, coaxial and fiber optic cables. This evolution has been driven by large increases in the number of users and the complexity and variety of the data transmitted over networks, resulting from:

•

the substantial growth in the Internet and its transformation from a text-based medium to a multimedia

•

platform containing pictures, video and sound;

•

the growth of wireless communications; and

•

the increased demand for remote network access and higher speed, higher bandwidth communication between LANs, MANs, and WANs.

The majority of installed communications systems were designed to transmit only voice communications and, therefore, are inadequate for the high-bandwidth transmission of both voice and data. As a result, new equipment is being deployed to augment existing transmission media and increase their bandwidth. Access to the public network is typically based on asynchronous technologies, such as T/E carrier over copper wire. The demand for greater bandwidth is driving a migration from lower-speed T1/E1 to higher-speed T3/E3 transmission rates. The T1/E1 standard permits the transmission of data at 1.5 Mbps /2.0 Mbps, and the T3/E3 standard permits the transmission of data at 45 Mbps/34 Mbps. The backbone of the public network is built on an optical fiber transmission medium that employs synchronous technologies such as SONET/SDH. Similar to the

utilization of faster transmission rates over copper wire, SONET/SDH protocols such as OC–3 (1.554 Mbps) and OC-12 (622 Mbps) are being upgraded to OC-48 (2.5 Gbps) and OC-192 (10 Gbps) to increase the bandwidth over a single optical fiber.

The market dynamics of the communications industry have changed significantly since 2000. Many carriers that spent billions of dollars building out their networks in 2000 were not successful in executing their business models and consequently were sold or declared bankruptcy. Incumbent telephone companies (such as AT&T, Sprint, MCI and Worldcom) and regional bell operating companies (such as BellSouth, Verizon and SBC) have substantially scaled back their capital expenditures in an effort to bring them in line with historical spending levels. This dynamic has severely impacted Exar’s customers, communications equipment OEMs, or original equipment manufacturers, many of which have undergone or are undergoing extensive restructuring to eliminate excess inventory and to reduce expenses by significantly downsizing their organizations. At the same time, despite these economic setbacks, the Company’s OEM customers are and will continue to come under tremendous pressure to develop new products for access and MAN markets that will enable carriers to leverage their current network infrastructure.

To address these evolving requirements of the communications industry, OEMs must develop and introduce increasingly sophisticated systems with fewer engineering resources. To achieve the performance and functionality required of these systems, communications OEMs are using increasingly complex communications ICs, which now account for a significant portion of the value-added proprietary content of these systems. As a result of the need for new equipment introductions, the reduction in their technical staff, the proliferation of transmission standards, and the difficulty of designing and producing communications ICs, equipment suppliers are increasingly outsourcing the design and production of the ICs incorporated into their systems.

These trends have created a significant opportunity for IC suppliers that can design cost-effective solutions for high-speed communications. The worldwide T/E carrier and related IC market is expected to grow over the next couple of years, with Salomon Smith Barney, or SSB, estimating that it will reach $922 million in potential revenue by 2004. In addition, SSB estimates that the SONET/SDH IC market will grow to $1 billion by 2004.

The key ICs contained in a typical communications system include physical-interface, access control, network processor, traffic manager and switch fabric devices. The physical-interface device consists of a transmitter and receiver that, when integrated, is called a transceiver. Transceivers interface with the physical transmission media, such as copper wire or optical fiber. Most of these high-speed, mixed-signal ICs convert parallel digital inputs into a single analog bit stream that is up to 32 times faster than the original signal. Transceivers therefore serve as a bridge between analog transmission media and the digital devices that process data. Access control circuits are digital ICs that format, or frame, the data, perform error checking and in some applications aggregate signals by mapping multiple lower speed data streams to a single higher speed stream. The traffic manager manages the transfer of data between the network processor and the switch fabric, which work together to shape, route and control the data.

Because physical-interface and access control ICs interface with the transmission media and are critical to increasing bandwidth, these ICs must offer high-speed and robust performance. Therefore, communications equipment OEMs seek IC suppliers that possess extensive analog and digital expertise to provide high-speed, mixed-signal solutions to bridge the analog physical world and the digital computing environment. This must be coupled with system-level expertise so that a supplier can quickly bring to market high-performance, highly reliable ICs with optimal feature sets.

The Exar Solution

Exar designs, develops and markets high-performance, high-bandwidth mixed-signal ICs for use in the worldwide communications infrastructure throughout the world. The Company’s analog and digital design expertise, combined with its systems understanding, enables the Company to provide physical-interface and

access control solutions for Access, MAN, and WAN communications equipment. Exar offers ICs based on the T/E carrier, ATM and SONET/SDH transmission standards. In addition, the Company provides solutions for the serial communications market and the video and imaging markets. Exar believes its products offer its customers the following benefits:

•

increased bandwidth through the integration of multiple channels on a single device;

•

reduced system noise/jitter to improve data integrity;

•

reduced overall system cost through the integration of multiple functions on a single device; and

•

accelerated time to market by allowing customers to focus on core competencies and outsource

•

standards-based solutions.

Key elements of the Company’s solution include:

Leading Analog and Mixed-Signal Design Expertise. Exar has over 30 years of experience in developing analog and mixed-signal ICs. As a result, the Company has developed a significant base of knowledge in these areas and a library of design elements. For example, the Company believes that it has particularly strong expertise in the design of high-speed, low-jitter phase lock loops, which are key elements in Exar’s mixed-signal transceiver, jitter attenuator and data aggregation mapper products. As a result, Exar can provide its customers with solutions that typically exceed standard specifications and allow them flexibility in designing other parts of their systems.

Broad Product Offerings. Exar offers a variety of physical-interface and access control products based on the T1/E1, T3/E3, ATM and SONET/SDH transmission standards. Exar offers a broad portfolio of multiple channel products for various transmission standards enabling its customers to minimize board space and overall cost in multi-port applications.

Comprehensive Solutions to Enhance System Integration. The combination of Exar’s design and system level expertise allows it to provide a solution that encompasses hardware, software and applications support. Exar believes that, by using its solutions, OEMs can efficiently integrate the Company’s devices into their systems, better leverage their development resources and reduce their time to market.

Compelling Price/Performance Solutions. The Company uses its systems expertise and its analog, digital and mixed-signal design techniques to architect high-performance products based on standard CMOS process technologies. Exar believes that these CMOS processes are proven, stable, predictable and able to meet its customers’ application speed and power/performance requirements at a lower price point than other semiconductor manufacturing processes.

Strategy

Exar’s objective is to be the leading provider of high-performance, high-bandwidth IC solutions for the worldwide communications infrastructure. To achieve this objective, Exar employs the following strategies:

Focus on Growing Market Share within the Communications Markets. Exar targets communications markets, including T/E carrier, ATM and SONET/SDH. The Company has built substantial expertise in the areas of analog and digital design, systems architecture and applications support. Exar believes that the integration of these capabilities enables the Company to develop solutions addressing the high-bandwidth requirements of communications systems OEMs. The Company’s broad product offerings support differentiated features, which the Company believes, will enable it to grow faster than the market by increasing its market share.

Leverage Analog and Mixed-Signal Design Expertise to Provide Integrated System Level Solutions. Utilizing Exar’s strong analog and mixed-signal design expertise, the Company integrates mixed-signal physical-interface devices with digital access control devices. The Company offers products that integrate transceivers

with jitter attenuator framers/ATM UNIs on a single IC. The Company’s data aggregation devices leverage its T/E carrier and SONET/SDH expertise, mapping multiple T3 data streams into a SONET/SDH stream. These configurations enable OEMs to use less board space and reduce their overall system cost.

Expand the Company’s Revenue Content Per System. Exar’s analog and mixed-signal design expertise enabled the Company to build what it believes to be a technological lead and a strong market position in T3/E3 transceivers. The Company intends to leverage this lead and its established customer relationships to capture design wins for its access control products, thereby increasing the Company’s overall revenue content per system.

Strengthen and Expand Strategic OEM Relationships. Exar’s customers include Alcatel, Cisco, Hewlett-Packard, Lucent, Nokia and Tellabs. To promote the early adoption of its solutions, the Company actively seeks collaborative relationships with strategic OEMs during product development. The Company believes that OEMs recognize the value of Exar’s early involvement because designing their system products in parallel with the Company’s development can accelerate time to market for their end products. In addition, Exar believes that collaborative relationships help the Company to obtain early design wins and to reduce the risk of market acceptance of its new products.

Leverage Broad Product Portfolio to Accelerate Communications Product Development. Exar believes it has developed a strong presence in the serial communications market, where the Company has leading industry customers and proven technological capabilities. Additionally, the Company markets a strong product portfolio in the video and imaging markets. The Company’s design expertise has enabled it to offer a diverse portfolio of both industry standard and proprietary serial communications products. The Company also has established important customer relationships for its high-performance, low-power video products and continues to work closely with key customers such as Hewlett-Packard for its imaging products. Exar’s sales to these markets provide the Company with resources to invest in and accelerate its communications product development.

Use Standard CMOS Process Technologies to Provide Compelling Price/Performance Solutions. Exar designs its products to be manufactured using standard CMOS processes. The Company believes that these processes are proven, stable and predictable and benefit from the extensive semiconductor-manufacturing infrastructure devoted to CMOS processes. Therefore, the Company believes that it can achieve a higher level of performance at a lower cost than competitors that use alternative processes.

Leverage Fabless Semiconductor Model. Exar has longstanding relationships with world-class third-party assembly, test and wafer foundries to manufacture the Company’s ICs. The Company’s fabless approach allows it to avoid substantial capital spending, obtain competitive pricing, reduce time to market, reduce technology and product risks, and facilitate the migration of the Company’s products to new process technologies, which reduce costs and optimize performance. By leveraging the fabless model, Exar can focus on its core competencies of IC design and product development.

Products

Exar designs, develops and markets high-performance, high-bandwidth physical-interface and access control solutions for the worldwide communications infrastructure. The Company’s current IC products for the communications market are designed to respond to the growing demand for high-speed networking equipment based on transmission standards such as T/E carrier, ATM and SONET/SDH. The Company also designs, develops and markets IC products that address the needs of the serial communications market and the video and imaging markets. Exar uses its design methodologies to develop products ranging from ASSPs (Application Specific Standard Products), designed for industry-wide applications, to semi-custom solutions for specific customer applications. These complementary products enable the Company to offer a range of solutions for its customers’ applications.

Communications

Exar’s products for T/E carrier, ATM and SONET/SDH applications include high-speed analog, digital and mixed-signal physical-interface and access control ICs. The physical-interface IC consists of a transmitter and receiver that, when integrated, is called a transceiver. Transceivers interface with the physical transmission media. Most of these high-speed, mixed-signal ICs convert parallel digital inputs into a single analog bit stream that is up to 32 times faster than the original signal. Access control circuits are digital circuits that format, or frame, the data, perform error checking and in some applications aggregate signals by mapping multiple lower speed data streams to a single higher speed stream. The figure below illustrates where the Company’s products are employed within networking equipment.

Types of Communications ICs Used in Networking Equipment

Exar’s communications products include transmitters and receivers, transceivers, jitter attenuators, framers, ATM UNIs (ATM User Network Interfaces) and data aggregation mappers. These products are used in networking equipment such as SONET/SDH multiplexers, private branch exchanges (PBX), central office switches, digital cross connects and multiservice provisioning platforms. The Company enhanced its physical-interface solution in 1999 with an integrated single chip transceiver. Subsequently, the Company announced two, three, four, six and twelve-channel versions of this transceiver that meet the same performance levels while requiring less board space and lower overall power in multi-port applications. In fiscal year 2001, the Company introduced its integrated, multi-channel jitter attenuator family, a proprietary solution that allows OEMs to meet difficult jitter transfer specifications while reducing overall system costs. Exar recently announced its next generation jitter attenuator capability, called “desynchronization,” which solves complex timing issues associated with mapping/demapping from SONET/SDH (synchronous) to T3/E3 (asynchronous) environments. The Company is embedding this new capability in its transceiver and data aggregation ICs. Exar’s access control products include framers, ATM UNIs and data aggregation mappers. These newer products are achieving greater market acceptance as Exar’s strong transceiver products have allowed it to compete for adjacent component opportunities. The Company also supplies a family of V.35 transceiver and multiprotocol products used for high-speed data transmission, primarily in networking equipment such as routers and bridges.

The following table describes some of the Company’s key communications products:



Product Description		Applications

OC-12 to 12 x T3/E3/STS-1 mapper OC-3 to 3x T3/E3/STS-1 mapper		SONET/SDH add/drop multiplexers, ATM switches, routers and digital cross connects

OC-48 framer		SONET/SDH add/drop multiplexers, ATM switches, routers and digital cross connects

T3/E3/STS-1 1, 2, 3, 4, 6, 12-channel transceivers and T3/E3/STS-1 1-channel receiver and transmitter		SONET/SDH multiplexers and digital cross connects

T3/E3/STS-1 1, 3-channel transceivers/jitter attenuators/desync		SONET/SDH multiplexers and digital cross connects

T3/E3 1, 3, 4-channel jitter attenuators		Multiplexers, switches and digital cross connects

T3/E3 1, 2, 3, 4, 6, 8-channel framers		Multiplexers and digital cross connects

T3/E3 1, 3, 4-channel ATM UNIs		ATM switches/routers/hubs

E1 4-channel transceiver and framer		Routers, internet access equipment, frame relay and ATM switches/routers/hubs

E1 multi-channel transceivers		Multiplexers, frame relay and ATM switches/routers/ hubs

T1/E1 1, 4, 8-channel SH and LH/SH transceivers		Routers, internet access equipment, frame relay and ATM switches/routers/hubs

T1 1, 4, 8-channel framers		Routers, internet access equipment, frame relay and ATM switches/routers/hubs

T1/E1 8-channel framer		Routers, internet access equipment, frame relay and ATM switches/routers/hubs

T1/E1 clock adaptors		Frame relay access devices and remote access servers

Multiprotocol serial interface		Multiplexers, access equipment and routers

V.35 serial interface		Multiplexers, access equipment and routers

The Company expects to introduce a number of new communications ICs in the fiscal year ending March 31, 2003 to provide an expanded line of T/E carrier products as well as SONET/SDH products. The T/E carrier products are expected to include multi-channel ICs that integrate transceivers and jitter attenuators (desync). Other T/E carrier products are expected to include multi-channel, multi-function ICs that integrate transceivers, jitter attenuators, framers and ATM UNIs. SONET/SDH product introductions are expected to focus on data aggregation, combining T3/E3 capability with OC-3 (155 Mbps), OC-12 (622 Mbps) or OC-48 (2.5Gbps) functions.

Serial Communications

UARTs convert data streams from parallel to serial, enabling a serial data stream to communicate with a central processing unit, or CPU. Exar sells its UART products to the remote access, data collection, industrial automation and handheld/mobile markets. Many of these products include high-performance features, such as automated flow control and large First-In First-Out, or FIFO, buffers. The Company has designed highly integrated quad (four channel) and octal (eight channel) UARTs with FIFO circuitry, which the Company believes are the de-facto industry standard for multi-channel FIFO UARTs used in multi-port networking applications.

The following table describes the Company’s key serial communications products:



Product Description		Applications

1, 2, 4-channel UART with 128 byte FIFO		Process control systems

2, 4, 8-channel PCI UART with 64 byte FIFO		PCI interface for network control management

1, 2, 4, 8-channel UART with 64 byte FIFO		Network management, remote access servers and point of sale systems, personal digital assistants and GPS

1, 2, 4-channel UART with 16 byte FIFO		Hub management, high-speed modems and PC I/O cards

2-channel UART with 8 byte FIFO		Process control systems, switches and serial port equipment

2-channel UART with 16 byte FIFO		Process control systems, switches and serial port equipment

2-channel UART		Serial port equipment

During the current fiscal year, the Company expects to expand its family of PCI multi-channel UARTs.

Video and Imaging

The video market is composed of several segments, including digital still cameras, or DSCs, PC video cameras, security cameras, camcorders and digital camcorders. Among these applications, one of the fastest growing segments is DSCs. To create images that are more comparable to film cameras and include features such as steady-shot and digital zoom, DSCs and digital camcorders are requiring higher-resolution and higher-speed data acquisition subsystems, also known as analog front ends, or AFEs, and analog-to-digital converters, or ADCs.

Exar supplies high-performance ADCs and integrated AFEs for products such as digital copiers, scanners or DSCs and multifunctional peripherals, or MFPs, which incorporate scanning, faxing and copying functions in a single integrated system. The Company uses advanced design techniques and process technologies to integrate low-power converter architectures with surrounding analog functions, thereby reducing system costs of the Company’s customers.

The following table describes some of the Company’s key video and imaging products:



Product Description		Applications

12bit/20 or 30 Msps AFEs		High speed scanners, DSCs, camcorders and video conferencing

10bit/18, 20 or 27 Msps AFEs		DSCs, camcorders and video conferencing

3-channel 12, 14 or 16bit/6 or 12 Msps AFEs		Scanners, MFPs and digital color copiers

10bit/20 or 40 Msps ADCs		High-end DSCs and broadcast video

8bit/6 Msps ADCs		Video boards, scanners and battery powered devices

8, 10 or 12 bit serial input DACs (digital-to-analog converter)		Voltage control and power control for wireless equipment

Sales and Customers

Exar markets its products in the Americas through 23 independent sales representatives and two independent, non-exclusive distributors, as well as through the Company’s own direct sales force. The Company currently has sales support offices in or near Atlanta, Boston, Chicago, Dallas, Los Angeles and Fremont, California. The Company is represented internationally by 24 sales representatives and distributors. In addition, the Company is represented in Europe by its wholly owned subsidiaries, Exar Ltd. and EXAR SARL, and in the Asia/Pacific region by its wholly owned subsidiaries, Exar Japan and Exar Taiwan.

The Company’s customers include the following, among others:


Communications	Serial Communications	Video and Imaging

• ADC Telecommunications, Inc. • Alcatel Alsthom S.A. • Cisco Systems, Inc. • Lucent Technologies, Inc. • Marconi Communications Plc. • NEC • Nokia Corporation • Siemens Corporation • Tellabs, Inc.	• AMI Inc. • Cisco Systems, Inc. • Digi International, Inc. • LM Ericsson Telephone Co. • Rose ElectronicsTellabs, Inc.	• Hewlett-Packard Company • Logitech International S.A. • Sharp Electronics Corp.

For the fiscal year ended March 31, 2002, one customer, Hewlett-Packard, accounted for 13.8% of the Company’s total sales. No other customer accounted for more than 10% of the Company’s total sales during this period. Sales through the domestic distribution channel accounted for 27.6% of revenue.

Manufacturing

Exar outsources all of its fabrication and assembly, as well as the majority of its testing operations. This fabless manufacturing model allows the Company to focus on its core competencies of product design and development.

The Company uses world-class independent wafer foundries, such as Chartered Semiconductor Manufacturing, or Chartered, and Taiwan Semiconductor Manufacturing Company, or TSMC. Chartered and TSMC manufacture all of the Company’s CMOS products. Chartered manufactures all of the Company’s BiCMOS (which combines bipolar and CMOS processes) products. Rohm Co. Ltd., which manufactured all of the Company’s bipolar products, ceased operations in December 2001. The Company does not intend to use an alternative supplier. Accordingly, consignment inventory is being held to support end of life requirements of our customers for these products. The Company does not have long term supply agreements with Chartered or TSMC. The majority of the Company’s current products are implemented in standard CMOS. The Company uses CMOS manufacturing processes to take advantage of that technology’s lower power consumption, cost-effectiveness, foundry availability and ever-increasing speed. Currently, all of the Company’s new product development is being implemented in CMOS.

Semiconductor wafers are usually shipped to the Company’s subcontractors in Asia for wafer test and assembly, where they are cut into individual dies and packaged. Independent contractors in Hong Kong, Indonesia and Singapore perform most of the Company’s assembly work. Following assembly, final test and quality assurance is performed either at the Company’s Fremont, California facility or at its subcontractors’ facilities in Asia. The Company conducts electrical testing of both wafers and packaged ICs. The combination of various functions makes the test process for analog and mixed-signal devices particularly difficult. Test operations require the programming, maintenance and use of sophisticated computer-based test systems and complex automatic handling systems.

Research and Development

Exar believes that the continued introduction of new products in its target markets is essential to its growth. The Company is focused on developing solutions addressing the high-bandwidth requirements of communications systems OEMs. As of March 31, 2002, the Company’s research and development staff consisted of 142 employees, 61 of who hold advanced engineering degrees. Over the next year, Exar will continue to seek to hire strategic technical and marketing personnel.

Competition

The semiconductor industry is intensely competitive and is characterized by rapid technological change and a history of price reductions as production efficiencies are achieved in successive generations of products. Although the market for analog and mixed-signal integrated circuits is generally characterized by longer product life cycles and less dramatic price reductions than the market for digital integrated circuits, the Company faces substantial competition in each market in which it participates. Competition in the Company’s markets is based principally on technical innovation, product features, timely introduction of new products, quality and reliability, performance, price, technical support and service. The Company believes that it competes favorably in all of these areas.

Because the IC markets are highly fragmented, the Company generally encounters different competitors in its various market areas. Competitors with respect to the Company’s communications products include Conexant Systems, Inc., PMC-Sierra and TranSwitch. In addition, the expansion of the Company’s communications product portfolio may in the future bring it into competition with other established communications IC companies, such as Applied Micro Circuits Corporation and Vitesse Semiconductor. Competitors in the Company’s other markets include Analog Devices Inc., Philips Electronics and Texas Instruments.

Backlog

Exar defines backlog to include OEM orders and distributor orders for which a delivery schedule has been specified for product shipment occurring primarily during the succeeding six months.

At March 31, 2002, Exar’s backlog was $15.1 million, compared with $18.1 million at March 31, 2001. Exar believes that the decrease in the Company’s backlog was due to a reduction in carrier capital expenditures and inventory accumulation within the communications product market. Additionally, the decrease in the backlog resulted from the shortened order-to-shipment time of the Company’s products. Over the past 12 to 18 months, the Company has experienced increased percentage of orders received that shipped in the same quarter. If this practice of short order-to-delivery times continue, the Company’s backlog may be less than in previous years and not a reliable indicator of future revenue.

Sales are made pursuant to purchase orders for current delivery of standard items or agreements covering purchases over a period of time, which are frequently subject to revision and cancellation. Lead times for the release of purchase orders depend on the scheduling practices of the individual customer and the rate of bookings varies from month to month. In addition, Exar’s distributor agreements generally permit the return of up to 10% of the purchases annually for purposes of stock rotation and also provide for credits to distributors in the event that Exar reduces the price of any product. Because of the possibility of changes in delivery schedules, quantities actually purchased, cancellations of orders, distributor returns or price reductions, Exar’s backlog as of any particular date may not be representative of actual sales for any succeeding six month period. Customers can cancel a significant portion, if not all, of their backlog at their discretion without substantial penalty.

Intellectual Property Rights

The Company has 100 patents issued and 26 patent applications pending in the U.S. The Company has 11 patents issued and 29 patent applications pending in various foreign countries. None of the Company’s key

domestic and foreign patents that have been issued will expire in the near future unless the Company chooses not to pay renewal fees. To protect its intellectual property, the Company also relies on a combination of mask work registrations, trademarks, copyrights, trade secrets, employee and third party nondisclosure agreements and licensing arrangements. The Company may enter into license agreements to gain access to externally developed products or technologies.

There can be no assurance that others will not independently develop substantially equivalent intellectual property or otherwise gain access to the Company’s trade secrets or intellectual property, or disclose such intellectual property or trade secrets, or that the Company can meaningfully protect its intellectual property. Furthermore, there can be no assurance that the Company’s pending patent applications or any future applications will be approved, or that any issued patents will provide the Company with competitive advantages, or will not be challenged by third parties, or that if challenged, will be found to be valid or enforceable, or that the patents of others will not have an adverse effect on the Company’s ability to do business. In addition, there can be no assurance that others will not independently develop similar products, duplicate the Company’s products, or design around any patents that may be issued to the Company.

The Company cannot be sure that its products or technologies do not infringe patents that may be granted in the future pursuant to pending patent applications or that the Company’s products do not infringe any patents or proprietary rights of third parties. Occasionally, the Company is informed by third parties of alleged patent infringement. In the event that any relevant claims of third-party patents are found to be valid and enforceable, the Company could be:

•

stop selling, incorporating or using its products that use the infringed intellectual property;

•

obtain a license to make, sell or use the relevant technology from the owner of the infringed intellectual property. Such license may not be available on commercially reasonable terms, it at all; and

•

redesign the Company’s products so as not to use the fringed intellectual property, which may not be technically or commercially feasible.

If the Company was forced to take any of the actions described above or defend against any claims from third parties, its business, financial condition and results of operations could be harmed.

Employees

As of March 31, 2002, the Company employed 290 full-time employees, with 142 in engineering and product development, 42 in operations, 59 in marketing and sales and 47 in administration. Of the 142 engineering and product development employees, 61 hold advanced degrees. The Company’s ability to attract, motivate and retain qualified personnel is essential to its continued success. None of the Company’s employees is represented by a collective bargaining agreement, nor has the Company ever experienced work stoppage. The Company believes its employee relations are good.

Facilities

Exar’s executive offices, marketing and sales, research and development, manufacturing and engineering operations are located in two buildings in Fremont, California that the Company owns, consisting of approximately 151,000 square feet. Additionally, the Company owns approximately 5.3 acres of undeveloped property adjacent to its headquarters, which is presently being held for future office expansion. The Company leases additional space for sales offices in or near Atlanta; Boston; Chicago; Dallas; Los Angeles; Kawasaki, Japan; and Velizy, France.

GLOSSARY: Exar uses a number of terms in this Annual Report which are familiar to industry participants but which some investors may not recognize. The Company has provided a glossary of some of these terms below.

Access Control Circuits:

An IC that formats or frames the data, performs error checking and in some applications aggregates signals by mapping multiple lower rate data streams to a single higher speed data stream.

Add/Drop Multiplexer (ADM):

A device at an intermediate point on a transmission line that enables new signals to come in and existing signals to go out.

Analog Integrated Circuits:

These semiconductor devices are used to electronically shape continuous real-world phenomena, such as sound waves, motion, heat, light and pressure. The electronic signals from analog ICs are typically translated into digital form and later converted back into analog forms that we can see or hear.

Application-Specific Standard Product (ASSP):

A device tailored for a specific application that is sold on the open market to multiple customers with similar requirements. Customers will often differentiate their end product with unique software and hardware features.

Asynchronous Transfer Mode (ATM):

A fast packet switching protocol by which short packets or cells containing data, voice or video signals are moved over networks at high-speed. This high-speed switching technology is used in both LAN and WAN applications. ATM is designed to take advantage of high-speed transmission media such as T3/E3 and SONET. Widely used as backbone technology in carrier networks and large enterprises.

Asynchronous Transmission:

Describes digital signals that are transmitted without precise clocking.

Backbone:

Backplane or bus which makes up the skeleton of a network.

Bandwidth:

Commonly defined as the volume of data that a transmission line can carry, measured in bits per second (bps). Traditional copper lines have the lowest bandwidth potential, while fiber optic lines have the highest.

A range of signal frequencies, measured in cycles per seconds or Hertz (Hz). Also refers to the speed at which data is transmitted, measured in bits per second (bps).

Broadband Communications:

Data transmission at speeds of greater than 1.5 Mbps.

Central Office (CO):

The main switching facility for a telephone company where larger telephone trunks are provisioned out in smaller lines to customers. Where fiber is connected to copper and vice versa.

Complementary Metal Oxide Semiconductor (CMOS):

Technology used to manufacture silicon integrated circuits.

Carrier:

An organization that provides communications services.

Copper:

Usually small gauge where two wires are twisted together, “twisted pair.” Limited in the amount of information it can carry. Typically associated with “last mile” transport.

Design Win:

A minimum order of 100 units which signifies acceptance of a device by an OEM for use in an end product.

Digital Cross Connect:

An electronic switching system that routes digital signals among multiple paths without demultiplexing them.

Digital ICs:

Within these devices, transistors are used to switch discrete digital signals that are represented in two states: on or off, or “1” or “0”. With today’s process technologies, millions of transistors can be integrated on a single chip, resulting in enormous computing power.

Frame Relay:

A packet-switching technology used to route frames of information within a WAN. Instead of leasing dedicated lines between all remote sites, frame relay allows virtual private networks to be established in which remote sites are connected to a central carrier, which routes data accordingly.

Gigabit (Gbps):

One billion bits per second.

Integrated Circuit (IC):

A single electronic device that contains thousands of previously separate (discrete) electronic components. An IC is produced on a small slice of silicon semiconductor material.

Kilobit (Kbps):

One thousand bits per second.

Local Area Network (LAN):

Local Area Network. A shared, private data communications network linking a variety of data devices such as workstations, computers and printers within an office or home environment, usually confined to a single building or cluster of buildings.

Megabit (Mbps):

One million bits per second.

Metropolitan Area Network (MAN):

A communications network that covers a geographic area, such as a city or suburb, or a series of LANs at multiple sites often interconnected by public facilities.

Mixed-Signal ICs:

Integrated circuits that combine analog and digital functions. Mixed-signal ICs are difficult to design because the analog and digital parts of the chip operate differently.

Multiplexing (Multiplexer or MUX):

An electronic or optical process that combines a large number of low-speed transmission lines into one high-speed line by splitting the total available bandwidth of the high-speed line into narrower channels. Demultiplexing is the reverse of multiplexing.

Network Processor:

An IC that executes programmed instructions, performs arithmetic and logical operations on signals, and controls input/output functions.

OC3-OC12-OC48-OC192:

Optical carrier, designating speed (155 Mbps, 622 Mbps, 2.5 Gbps, 10 Gbps). Designates SONET speed.

Packet:

Any logical block of data sent over a network; it contains a header consisting of control information such as sender, receiver, and error-control data, as well as the message itself. May be fixed or variable length.

Phase Lock Loop (PLL):

Analog or digital function that regulates the critical timing function in telecommunication and computer ICs.

PHY:

Physical layer—Level 1 on the OSI (Open Space Interconnect) Model. The connection layer, or where the “wire” connects to the equipment.

Physical-Interface:

An IC that receives analog signals from transmission media and converts it to a digital signal for further processing for incoming signals. It converts digital signals to analog for transmission across the transmission media for outgoing signals.

Router:

A device that connects multiple LANs, or the bridge between a LAN and the public network.

Synchronous Digital Hierarchy (SDH):

An international standard for synchronous data transmission over fiber; SDH was adapted from SONET, which is employed in North America.

Synchronous Optical Network (SONET):

The North America standard for high-speed synchronous data transmission over glass fiber, optical media. Rates are measured in optical carrier (OC) units. For example, OC-3 equals 155 Mbps, OC-12 equals 622 Mbps and OC-48 equals 2.5 Gbps. SONET backbones are widely used to aggregate lower-speed T1 and T3 lines. The international equivalent is SDH.

Synchronous Transport Signal (STS-x):

Electronic equivalent of SONET—OC-level, x designating the speed.

Switch Fabric Devices:

An IC that incorporates a series of point to point interconnections that are used to overcome bandwidth bottlenecks in traditional bus structures. Connection end points are joined through a cross-connected switching matrix resembling a woven piece of cloth, hence the name fabric. Any point may be combined with any other point to produce a powerful combination of switched end points.

T1/E1:

T1 is a dedicated digital transmission link with a capacity of 1.554 Mbps provided by telephone companies. Widely used for private networks as well as interconnections between an organization’s PBX or LAN and telephone companies. A T1 line can normally accommodate 24 voice conversations (channels), each one digitized at 64 Kbps and one 8 Kbps channel for signaling and control. E1 is the European designation for T1 and has a capacity of approximately 2 Mbps.

T3/E3:

A digital transmission link capable of transmission speeds of 45 Mbps. A T3 line can normally accommodate 672 voice conversations. E3 is the European designation for T3.

Traffic Manager:

An IC which manages the transfer of data among numerous channels and elements such as the network processor and switch fabric, which work together to shape, route and control the data.

Transmission Media:

The physical transmission path along which a signal propagates, such as a twisted copper pair, coaxial cable or optical fiber cable.

Wide Area Network (WANs):

A shared network of computers and data devices that spans a large geographical area, such as the Internet, normally consisting of multiple LANs or MANs.

MANAGEMENT

The names of the Company’s executive officers and directors, and their ages as of May 31, 2002, are as follows:

Name	Age	Position
Donald L. Ciffone, Jr.	46	Chairman of the Board, Chief Executive Officer and President
Michael Class	44	Vice President, Worldwide Sales
Roubik Gregorian	52	Executive Vice President/General Manager and Chief Technology Officer, Communications Division
Ronald W. Guire	53	Executive Vice President, Chief Financial Officer, Assistant Secretary and Director
Susan J. Hardman	40	Vice President, Corporate Marketing
Thomas R. Melendrez	48	Corporate Vice President, General Counsel and Secretary
Stephen W. Michael	55	Vice President, Operations and Reliability & Quality Assurance
Raimon L. Conlisk	79	Vice Chairman and Director
Frank P. Carrubba	64	Director
James E. Dykes	64	Director
Richard Previte	67	Director

Donald L. Ciffone, Jr. joined the Company as President and Chief Executive Officer in October 1996 and was appointed a Director at that time. Mr. Ciffone was appointed Chairman of the Board in April 2002. From August 1996 to October 1996, Mr. Ciffone was Executive Vice President of Toshiba America, the U.S. semiconductor subsidiary of Toshiba Semiconductor. Prior to joining Toshiba, he served from 1991 to 1996 in a variety of senior management positions, including Senior Vice President of the VLSI Product Divisions, at VLSI Technology, Inc. From 1978 to 1991, Mr. Ciffone held a variety of marketing and operations positions at National Semiconductor, Inc. Mr. Ciffone is on the board of Internet Machines Corporation, a privately held communications IC company focused on network processor, traffic manager and switch fabric devices. Mr. Ciffone holds a B.A. from San Jose State University and an M.B.A. from Santa Clara University.

Michael Class joined the Company as Director of Western Area Sales in 1997. In January 1998, he was promoted to the position of Vice President, North American/European Sales and was promoted to Vice President, Worldwide Sales in July 1999. Mr. Class has over 20 years of experience in the semiconductor industry, most recently with IC Works, Inc. as Area Sales Manager for the Western U.S. and Canada. Prior to joining IC Works, Mr. Class held various sales management positions with Intel Corporation and VLSI from 1979 to 1995. He holds a B.S. in Electrical Engineering from Lehigh University and an M.B.A. from LaSalle University.

Roubik Gregorian joined the Company in March 1995 as Vice President, Startech Division, when the Company acquired Startech Semiconductor, Inc., where he served as President. He was appointed Chief Technology Officer and Vice President of the Communications Division in June 1996, and to his current position as Chief Technology Officer, Senior Vice President/General Manager, Communications Division, in June 1998. Dr. Gregorian was promoted to Executive Vice President/General Manager, Communications division in May 2002. Prior to joining Startech in 1994, Dr. Gregorian was Vice President of Research and Development and Chief Technology Officer for Sierra Semiconductor, Inc. Dr. Gregorian has been issued 24 patents and received his M.S.E.E. and Ph.D. in Electrical Engineering from the University of California at Los Angeles, as well as an M.S.E.E. from Tehran University.

Ronald W. Guire joined the Company in July 1984 and has been a Director since June 1985. He has served as Chief Financial Officer since May 1985, and Executive Vice President since July 1995. Mr. Guire was Chairman of the Board of Xetel Corporation, an electronics contract manufacturer until May 2002. Mr. Guire was a partner in the certified public accounting firm of Graubart & Co. from 1979 until he joined Exar in July 1984. Mr. Guire holds a B.S. in Accounting from California College of Commerce.

Susan J. Hardman joined the Company in February 1997 and became Vice President, Corporate Marketing in February 2000. Prior to this position, she served as Senior Director of Business Development as well as Director of Marketing for the Company’s communications products. Ms. Hardman has over 18 years experience in the semiconductor industry. From 1989 to 1997, Ms. Hardman was with VLSI in a variety of management positions, most recently as Director of Product Marketing for VLSI’s networking products division. From 1983 to 1989, she was with Motorola holding a variety of engineering roles. Ms. Hardman holds a B.S. in Chemical Engineering from Purdue University and an M.B.A. from the University of Phoenix.

Thomas R. Melendrez joined the Company in April 1986 as Corporate Attorney. He was promoted to Director, Legal Affairs in July 1991, and again to Corporate Vice President, Legal Affairs in March 1993. In March 1996, Mr. Melendrez was promoted to his current position of Corporate Vice President, General Counsel. In June 2001, Mr. Melendrez was appointed Secretary of the Company. Mr. Melendrez has over 20 years legal experience in the semiconductor and related industries. He received a B.A. from the University of Notre Dame, a J.D. from the University of San Francisco, and an M.B.A. from Pepperdine University.

Stephen W. Michael joined Exar as Vice President New Market Development in September 1992. In July 1995, he was appointed to his current position of Vice President, Operations and Reliability & Quality Assurance. Mr. Michael has over 25 years of semiconductor industry experience, most recently as Vice President and General Manager, Analog and Custom Products with Catalyst Semiconductor. He joined Catalyst in 1987 and served in various senior positions.

Raimon L. Conlisk joined the Company as a Director in August 1985, was appointed Vice Chairman of the Board in August 1990, served as Chairman of the Board from August 1994 to April 2002, and is currently serving as Vice Chairman. Mr. Conlisk has also served as a director since 1991, and in December 1997 was appointed Chairman of the Board of SBE, Inc., a manufacturer of communications and computer products. From 1977 to 1999, Mr. Conlisk was President of Conlisk Associates, a management consulting firm serving high-technology companies in the United States and foreign countries. From 1991 to 1998, Mr. Conlisk served as a Director of Xetel Corporation, a contract manufacturer of electronic equipment. Mr. Conlisk was also President from 1984 to 1989, a Director from 1970, and Chairman from 1989 until retirement in June 1990, of Quantic Industries, Inc., a privately held manufacturer of electronic systems. From 1970 to 1973, and from 1987 to 1990, Mr. Conlisk served as a Director of the American Electronics Association.

Frank P. Carrubba joined the Company as a Director in August 1998. Dr. Carrubba served as Executive Vice President and Chief Technical Officer of Royal Philips Electronics N.V., headquartered in Eindhoven, The Netherlands, from 1991 to 1997. From 1982 to 1991, Dr. Carrubba was with the Hewlett-Packard Company, where he was a member of the Group Management Committee and was Director of Hewlett-Packard Laboratories. Prior to joining Hewlett-Packard, he spent 22 years as a member of the technical staff at IBM Corporation’s Thomas J. Watson Research Laboratory in Yorktown Heights, New York. Dr. Carrubba was one of the original designers of the RISC Architecture, for which he was named “Inventor of the Year” by the Intellectual Property Owners in Washington, D.C. in 1992. Dr. Carrubba is also a Director of Coherent, Inc., a global leader in the design, manufacture and sale of lasers, and Gyration, Inc., creators of in-air cordless mice and controllers for games, presentations and virtual reality. In June 2002, Dr. Carrubba was appointed Chairman of the Board of Accerra Corporation, a provider of secure online services for business communications located in Santa Rosa, California.

James E. Dykes joined the Company as a Director in May 1994. Mr. Dykes served as President and CEO of the Signetics division of North American Philips Corporation, a manufacturer of industrial and consumer electronics, from 1989 to 1993, and from 1987 to 1988, as President and CEO of TSMC, a semiconductor foundry in Taiwan. Prior to joining TSMC, Mr. Dykes held various management positions with other semiconductor and related companies, including General Electric Company, a diversified international manufacturer of defense, electrical and other products, and Harris Semiconductor, Inc., a manufacturer of semiconductors and integrated circuits. From August 1994 to June 1997, Mr. Dykes served as President and

Chief Operating Officer of Intellon Corp., a wireless network communications company. From July 1997 to July 1998, Mr. Dykes served as Executive Vice President, Corporate Development, of the Thomas Group, Inc., a management services company. Mr. Dykes also is a Director of the Thomas Group, Inc., Cree Research, Inc., a silicon carbide materials and electronics company, and Thesus Logic, Inc., a privately held semiconductor.

Richard Previte joined the Company as a Director in October 1999. He was a director of Advanced Micro Devices, or AMD, from 1990 to April 2000, and Vice Chairman from 1999 to April 2000. Additionally, Mr. Previte served as Chairman of the Board of Vantis Corporation, a subsidiary of AMD, from 1997 to June 1999, and acted as Chief Executive Officer from February 1999 to June 1999. Mr. Previte served as President of AMD from 1990 to 1999, Executive Vice President and Chief Operating Officer from 1989 to 1990, and Chief Financial Officer and Treasurer from 1969 to 1989. Most recently he was Chief Executive Officer and Chairman of the Board of MarketFusion, Inc., from January 2000 to April 2002.

ITEM 2. PROPERTIES

The Company’s corporate headquarters are located in Fremont, California, consisting of approximately 151,000 square feet. The land and building are owned by the Company and house Exar’s principal administrative, test, engineering, marketing, customer service and sales departments. Additionally, the Company owns approximately 5.3 acres of undeveloped property adjacent to its headquarters, which is presently being held for future office expansion. The Company leases additional space for sales offices in or near Atlanta; Boston; Chicago; Dallas; Los Angeles; Kawasaki, Japan; and Velizy, France.

ITEM 3. LEGAL PROCEEDINGS

The Company is not currently a party to any material legal proceedings.

ITEM 4. SUBMISSION OF MATTERS TO A VOTE OF SECURITY HOLDERS

During the fourth quarter of fiscal 2002, no matter was submitted to a vote of security holders through the solicitation of proxies or otherwise.

PART II

ITEM 5.

MARKET FOR REGISTRANT’S COMMON STOCK AND RELATED STOCKHOLDER MATTERS

The Common Stock of Exar is traded on The Nasdaq Stock Market, Inc. under the symbol “EXAR.” The following table sets forth the range of high and low sales prices for the Company’s Common Stock for the periods indicated, as reported by Nasdaq and as adjusted for a three-for-two stock split effected on February 15, 2000 and a two-for-one stock split effected on October 19, 2000. The listed quotations represent inter-dealer prices without retail markups, markdowns or commissions.

The Company has never paid cash dividends on its Common Stock and presently intends to continue this policy in order to retain earnings for use in its business. The Company had approximately 214 stockholders on record as of May 31, 2002. The Company believes it has in excess of 8,570 beneficial stockholders. The last sales price for Exar’s Common Stock, as reported by Nasdaq on May 31, 2002 was $21.57 per share.

	Common Stock Prices
	High		Low
FISCAL 2002
Quarter ended March 31, 2002	$	25.11	$	17.82
Quarter ended December 31, 2001	$	25.70