Electronics Guide

Principles of Spectrum Spreading

An ordinary communication signal occupies a bandwidth roughly proportional to its data rate. Spread-spectrum systems break this proportionality by modulating the data with a spreading code whose rate, measured in chips per second, greatly exceeds the data rate. The result is a transmitted signal whose bandwidth is set by the chip rate rather than the data rate, often hundreds or thousands of times wider than a conventional signal carrying the same information.

Two conditions define a true spread-spectrum system. First, the transmitted bandwidth must be much greater than the bandwidth of the information. Second, the spreading must be governed by a code independent of the data, known to the intended receiver but appearing random to others. These conditions distinguish spread spectrum from wideband schemes, such as wideband frequency modulation, that expand bandwidth as a byproduct of the modulation itself rather than through an independent code.

The benefit of spreading emerges at the receiver. Correlating the received signal with a synchronized replica of the spreading code despreads the wanted signal, restoring it to its narrow information bandwidth, while simultaneously spreading any narrowband interference across the wide band. A narrowband filter following the despreader then passes most of the wanted signal energy but only a small fraction of the spread interference, producing a gain in signal-to-interference ratio that is the central advantage of the technique.

Direct-Sequence Spread Spectrum

Direct-sequence spread spectrum (DSSS) spreads the signal by multiplying the data stream directly with a high-rate pseudonoise code. Each data bit is represented by a fixed sequence of code chips, so the transmitted waveform changes at the chip rate. The wide bandwidth follows directly from the rapid chip transitions, and the structure of the code determines the spectral and correlation properties of the signal.

Spreading and Despreading

At the transmitter, the data signal and the spreading code, both represented as sequences of plus and minus one, multiply together before modulating the carrier. Because the code chips change far faster than the data bits, the product occupies the wide bandwidth of the code. At the receiver, multiplication by a time-aligned copy of the same code reverses the operation. Since the product of the code with itself equals one at every chip, the wanted data emerge intact, restored to their original bandwidth. Any signal not carrying the matching code, including interference and the signals of other users, is instead multiplied by the code for the first time at the receiver and spread across the wide band.

Resistance to Narrowband Interference

The despreading operation reveals why direct-sequence systems resist narrowband interference. A strong interferer that occupies a small part of the band passes through the receiver's despreading multiplier and emerges spread across the full code bandwidth, with its power density correspondingly reduced. The narrowband filter after despreading rejects most of this spread energy. The wanted signal, by contrast, is concentrated by despreading and passes through the filter with little loss. The receiver thereby suppresses interference in proportion to the ratio of the spread bandwidth to the information bandwidth.

Multipath Behavior

The sharp autocorrelation of a good spreading code allows a direct-sequence receiver to distinguish signal replicas that arrive by different paths and that are delayed by more than one chip period. A rake receiver exploits this property by assigning separate correlators, or fingers, to the strongest delayed copies and combining them constructively. Rather than suffering from multipath, such a receiver gains diversity from it, which is a notable advantage in mobile environments.

Frequency-Hopping Spread Spectrum

Frequency-hopping spread spectrum (FHSS) spreads the signal in a different manner. Instead of widening the instantaneous bandwidth, it transmits a relatively narrowband signal whose carrier frequency changes rapidly according to a pseudorandom hopping pattern. Over time the signal visits many frequencies across a wide band, so the average occupied bandwidth is large even though the bandwidth at any instant is modest. The hopping sequence, derived from a pseudonoise code, is known to the intended receiver, which retunes in step to follow the signal.

Fast and Slow Hopping

Frequency-hopping systems are classified by the relationship between the hop rate and the symbol rate. In slow hopping, several symbols are transmitted on each frequency before the next hop. In fast hopping, the carrier hops several times within a single symbol, which provides additional protection because an interferer must corrupt every hop of a symbol to destroy it. The choice involves a trade-off between the complexity and stability of the rapidly retuning synthesizer and the degree of robustness obtained.

Interference and Jamming Behavior

A frequency-hopping signal avoids a narrowband interferer or a jammer by spending only a fraction of its time on any single frequency. When the signal happens to hop onto an occupied frequency, the affected symbols may be corrupted, but error-correcting codes and interleaving readily recover from such intermittent losses. Against a partial-band jammer that concentrates its power on a portion of the band, frequency hopping forces the jammer to choose between covering the whole band weakly or part of it strongly, and coding defeats either choice. This graceful behavior in hostile environments accounts for the prevalence of frequency hopping in tactical military radios and in the adaptive hopping used by short-range commercial systems to coexist with other devices.

Processing Gain

Processing gain quantifies the advantage that spreading confers against interference and noise. It equals the ratio of the spread bandwidth to the information bandwidth, or equivalently the ratio of the chip rate to the data rate in a direct-sequence system. Expressed in decibels, the processing gain is ten times the base-ten logarithm of this ratio. A system that spreads a signal by a factor of one thousand achieves a processing gain of thirty decibels, meaning that despreading improves the signal-to-interference ratio by that amount relative to the ratio present in the wide channel.

Processing gain does not improve performance against broadband thermal noise, because such noise already fills the channel and is unaffected by spreading. Its benefit applies to interference that the despreader spreads, including narrowband jammers, the signals of other spread-spectrum users sharing the band, and intentional jamming. The related concept of jamming margin expresses how much stronger than the wanted signal an interferer may be while the receiver still operates, and it equals the processing gain reduced by the required signal-to-noise ratio and by implementation losses. A large processing gain therefore translates directly into the capacity to operate beneath strong interference, which is the defining capability of spread spectrum.

Pseudonoise Sequences

The spreading code is the heart of any spread-spectrum system, and its properties determine how well the system performs. Such codes are called pseudonoise or pseudorandom sequences because they appear noiselike and statistically random, yet they are generated deterministically and can be reproduced exactly by an authorized receiver.

Desired Properties

A good spreading code possesses three properties. It has a sharp autocorrelation, meaning that the code correlates strongly with itself only when perfectly aligned and weakly at all other shifts, which enables timing acquisition and multipath resolution. It has low cross-correlation with the codes assigned to other users, which limits the mutual interference among signals sharing a band. And it has balance and randomness in the distribution of its chips, which produces a smooth, noiselike spectrum.

Common Code Families

Maximal-length sequences, generated by linear feedback shift registers, achieve nearly ideal autocorrelation and are simple to produce, but the number of distinct sequences of a given length is limited and their mutual cross-correlation is not always small. Gold codes, formed by combining pairs of maximal-length sequences, provide large families of codes with bounded cross-correlation, which makes them well suited to systems in which many users transmit simultaneously, including the satellite navigation signals of the Global Positioning System. Kasami sequences offer similarly favorable cross-correlation for large code sets. Walsh functions, which are mutually orthogonal when perfectly synchronized, serve to separate channels in synchronous systems such as the downlink of certain cellular standards, where all signals share a common timing reference.

Code-Division Multiple Access

Spread spectrum enables a powerful method of sharing a channel among many users. In code-division multiple access (CDMA), every user transmits over the same frequency band at the same time, distinguished not by frequency or time slot but by a unique spreading code. A receiver despreads the signal with the code of the desired user, concentrating that user's energy while leaving the other users spread as noiselike interference. The low cross-correlation among the assigned codes keeps this multiple-access interference manageable, so that many users coexist in the same band.

Capacity and Interference

The capacity of a CDMA system is limited not by a fixed number of channels but by the aggregate interference that all active users present to one another. Each additional user raises the interference floor slightly, so capacity degrades gracefully as load increases rather than collapsing at a hard limit. This interference-limited behavior makes power control essential. Because a strong signal from a nearby transmitter would otherwise overwhelm the despread signal of a distant one, the system continuously adjusts each transmitter's power so that all signals arrive at the receiver at comparable strength. This near-far problem and its solution through tight power control are defining characteristics of practical CDMA.

Role in Cellular Systems

Code-division multiple access shaped a generation of cellular technology. Second-generation systems based on the IS-95 standard and the third-generation Universal Mobile Telecommunications System employed wideband CDMA as their air interface, exploiting its soft capacity, frequency reuse across every cell, and rake reception of multipath. Although later cellular generations adopted orthogonal frequency-division techniques for their main data channels, the principles of spreading and code-division access remain influential and persist in control channels and in satellite systems.

Jam Resistance and Low Probability of Interception

The military origins of spread spectrum reflect two properties that remain among its most valued. Jam resistance follows from processing gain. An adversary attempting to deny communication must inject enough power into the wide spread bandwidth to overcome the wanted signal after despreading, and the processing gain forces the jammer to expend far more power than a conventional signal would require. Frequency hopping compounds the difficulty by denying the jammer knowledge of where the signal will appear next, while direct-sequence spreading buries the signal beneath a wideband waveform that resists narrowband jamming.

Low probability of interception (LPI) and low probability of detection (LPD) follow from spreading the signal energy thinly across a wide band. Because the transmitted power is distributed over a bandwidth far larger than the information requires, the power spectral density of the signal can fall below the noise floor of an unintended receiver, rendering the transmission difficult to detect, let alone demodulate, without knowledge of the spreading code. A direct-sequence signal can be made to appear as a slight rise in the background noise, and a frequency-hopping signal presents only brief, scattered bursts to an observer who does not know the hopping pattern. These properties make spread spectrum attractive for covert and secure communication, and the secrecy of the spreading code provides an additional layer of access control, though it does not by itself substitute for cryptographic protection of the message.

Synchronization

A spread-spectrum receiver cannot recover any information until it aligns its local code with the incoming signal, because despreading requires the local and received codes to match in both code phase and, for frequency hopping, hop timing. Synchronization is therefore a critical and often demanding function, conventionally divided into acquisition and tracking.

Acquisition

Acquisition is the initial process of bringing the local code into coarse alignment with the received code, typically within a fraction of a chip. Because the receiver does not know the correct code phase in advance, it must search across the possible alignments until correlation indicates a match. A serial search tests candidate phases one at a time and is simple but slow, whereas parallel correlators or transform-based methods test many phases at once to acquire more quickly at the cost of greater complexity. The size of the search, and hence the acquisition time, grows with the length of the code and the uncertainty in timing and frequency.

Tracking

Once acquisition achieves coarse alignment, tracking maintains fine synchronization as the relative timing drifts because of clock differences and motion. A delay-locked loop is the usual mechanism, comparing correlations of the received signal with slightly early and late copies of the code and adjusting the local code timing to keep it centered on the incoming signal. A parallel carrier-tracking loop maintains frequency and phase alignment. The precision of code tracking is what allows spread-spectrum systems to measure propagation delay accurately, which is the basis of their ranging and positioning capability.

Applications

Spread-spectrum techniques appear across a broad range of systems, civil and military, that exploit one or more of their distinctive properties.

Satellite Navigation

Global navigation satellite systems, including the Global Positioning System, rely on direct-sequence spread spectrum. Each satellite broadcasts a unique spreading code, and a receiver measures its position by correlating against these codes to determine the propagation delay, and hence the distance, from several satellites at once. The sharp autocorrelation of the codes provides the precise timing that positioning demands, and code division allows all satellites to share the same frequency.

Cellular and Wireless Systems

Code-division multiple access served as the foundation of major second- and third-generation cellular networks, and spreading concepts persist in modern systems. Short-range wireless technologies apply spread spectrum widely. Frequency-hopping underlies the adaptive hopping that allows certain personal-area networks to coexist in crowded unlicensed bands, and direct-sequence spreading appeared in early wireless local area network standards. These commercial uses exploit spread spectrum chiefly for its resistance to interference and its ability to share unlicensed spectrum gracefully.

Military and Specialized Links

Military communication systems employ spread spectrum for jam resistance and for low probability of interception and detection, combining frequency hopping and direct-sequence spreading in tactical radios and secure data links. Beyond communication, the precise ranging that spread spectrum permits supports radar and distance-measurement systems, and the technique finds use in any application that benefits from operating beneath interference or beneath the notice of an unintended observer.

Summary

Spread-spectrum communication trades bandwidth for robustness by spreading a signal with a pseudonoise code independent of the data and despreading it with a synchronized replica at the receiver. Direct-sequence systems spread by multiplying the data with a high-rate code, widening the instantaneous bandwidth and resisting narrowband interference, while frequency-hopping systems spread by retuning a narrowband carrier across many frequencies in a pseudorandom pattern, evading jammers and interference. Processing gain, the ratio of spread to information bandwidth, measures the resulting advantage against interference and sets the jamming margin. Pseudonoise sequences with sharp autocorrelation and low cross-correlation, such as maximal-length, Gold, and Walsh codes, make spreading and code-division access possible. Code-division multiple access lets many users share a band, limited by mutual interference and dependent on power control. Spread spectrum confers jam resistance and a low probability of interception and detection, demands careful acquisition and tracking for synchronization, and underlies satellite navigation, cellular and short-range wireless systems, and secure military links.

Related Topics

• Cellular Mobile Systems
• Wireless Local Area Networks
• Bluetooth and BLE
• Satellite Communication Systems
• Ultra-Wideband and Precision Ranging
• Modulation and Signal Processing
• Radio Frequency Systems