PRODUCT SEARCH

AIRLINX

NEWS

Fri, 24 Oct 14 05:16 pm EST    Languages: Español  Français  Português
Extranet Login | Site Map | Contact Us

Our Coordinates: Lat 42.75 N, Long 71.85 W  


Home Radio & Technology Spread Spectrum Summary

Spread Spectrum Basic Summary

Quite simply, spread spectrum is a coding technique for digital transmission. It was originally developed for the military under a veil of secrecy. The purpose of coding is to transform an information signal so that it looks more like noise. Noise has a flat uniform spectrum with no coherent peaks and can be reduced or eliminated by filtering. The spread spectrum coding technique modifies the signal spectrum to spread it out and increase its bandwidth. The new "spread" signal has a lower power density, but the same total power. 

The expanded transmitter bandwidth minimizes interference to others because of its low power density. In the receiver, the incoming signal is decoded, and the decoding operation provides resistance to interference and multipath fading. 

Usually, spread spectrum is implemented for two processes -- frequency hopping and direct sequence. 

Frequency Hopping 

In frequency hopping systems, the carrier frequency of the transmitter abruptly changes ("or hops") in accordance with an apparently random pattern. This pattern is in fact a pseudo-random code sequence. The order of the frequencies selected by the transmitter is taken from a predetermined set as dictated by the code sequence. The receiver tracks these changes and produces a constant IF signal. Interfering signals are not tracked. Therefore they only occasionally fall within the IF bandwidth of the receiver. 

Fast frequency hopping systems change frequency at a significantly higher rate than the information rate. Slow frequency hopping systems change frequency at a rate comparable with (or slower than) the information rate. 

Direct Sequence 

In direct sequence systems, the carrier phase of the transmitter abruptly changes in accordance with a pseudo-random code sequence. This process is generally achieved by multiplying the digital information signal with a spreading code, also known as a chip sequence. The chip sequence has a much faster data rate than the information signal and so expands or spreads the signal bandwidth beyond the original bandwidth occupied by just the information signal. The term chips are used to distinguish the shorter coded bits from the longer uncoded bits of the information signal. 

At the receiver, the information signal is recovered by remultiplying with a locally generated replica of the spreading code. The multiplication process can be accomplished by an exclusive OR gate, and in the receiver effectively compresses the the spread signal back to its original unspread bandwidth. 

The amount of spreading, for direct sequence, is dependent on the ratio of "chips per bit". Also, the same chip sequence must be used in the receiver as in the transmitter to recover the information. 

Interfering signals are reduced by the process gain of the receiver. They are spread beyond the desired information bandwidth by the second multiplication process (in the receiver) and then removed by filtering. 

Power Density

Power density is measured as the power in a given bandwidth, for example, dBm in 3 kHz. It is always a maximum in an unmodulated carrier (CW). All the RF output power is present in a very narrow bandwidth around the CW carrier. Modulated signals have different power densities, as seen by measuring the RF output power in a given bandwidth across the RF channel. Spread spectrum signals attempt to produce a very uniform (flat) power density with no coherent peaks by using pseudo-random code sequences. The closer the code is to being completely random, the more uniform the power density will become. Spread spectrum signals never achieve a completely uniform power density and will always exhibit a fine line-structured spectrum. The frequency separation of the line spectra is reduced by increasing the code repetition rate with a faster chip rate or a longer code.

Codes

Pseudo-random spreading codes have a fixed length. After a fixed number of chips (the code length) they repeat themselves exactly. Codes may be formed using a shift register with feedback taps. A common useful series of codes (maximal length codes) 127 chips long may be formed using a 7-bit shift register. 

Good codes have a low cross-correlation response. This results in minimum interference between users, especially when signals are synchronized. A receiver that is set to use one particular code can only be reached by the transmitter sending the same code. Cross-correlation is the measure of agreement between two different codes. It can be calculated by determining the number of agreements minus the number of disagreements when the codes are compared chip by chip, while one code is shifted one chip at a time.

Good codes also have a high auto-correlation peak, when exactly lined up, which minimizes false synchronization. Auto-correlation is the same as cross-correlation, except the code is compared against itself, with a relative shift of one chip at a time. 

Synchronization 

The most difficult part of designing a spread spectrum radio is to ensure fast reliable and synchronization in the receiver. The receiver must correlate the incoming signal and then demodulate it. The correlator removes the spreading code and the demodulator recovers the information at baseband. Both must be synchronous with the transmitted signal and usually lock up to the incoming signal and track it. Acquisition time is the period taken to lock up the receiver from a cold start and is an important measure of the receiver's performance. Other measures include the ability to synchronize in the presence of interference and/or thermal noise and to remain synchronized over long periods.