Radio Performance Concepts

Sensitivity
Sensitivity is a concept that refers to the minimum usable signal level at the receiver. In the case of the GSM handset, the sensitivity level is around -100 dBm, while for the BTS this level reaches approximately -106 dBm. This difference appears due to the higher quality electronics used in the basestation.

Signal-to-Noise Ratio (SNR)
Signal-to-noise ratio is a method that calculates the ratio of the desired signal power to the background noise power and is expressed in decibels.

Any ratio higher than 1:1, or greater than 1 dB, means that there is more signal than noise.

Bit Error Rates
In digital transmissions, bit errors represent the number of alerted bits entering a data stream through a communication channel. Bits can be distorted by noise, interference or bit synchronization errors.

The bit error rate is the ratio between the bit errors and the total number of transferred bits during a particular time.

A receiver's bit error rate can be affected by noise, interference or multipath fading. To improve it, one must use a more robust modulation scheme that would lead to transmitting a stronger signal, or to apply error correction coding schemes.

There are two types of bit error rates:
 * transmission bit error rate – the number of incorrect bits divided by the total number of transmitted bits
 * information bit error rate – the number of decoded bits that remain incorrect after applying error correction schemes divided by the total number of decoded bits

Usually, the transmission bit error rate is greater than the information bit error rate.

Error Correction Coding
Error correction coding is a method of controlling errors during data transmissions over noisy channels of communication, using redundant message encoding. The main purpose of error correction coding is to allow the receiver to spot a number of errors during the message transmission and to correct them without retransmission and, as such, without a reverse channel for the retransmission.

Thermal Noise
Thermal noise is the electronic noise produced by the natural motion of the electrons in a receiver's atoms. It largely affects the receiver's quality. If your receiver is made out of atoms, then it will automatically produce thermal noise.

This phenomenon is directly proportional to resistance and temperature, therefore, the lower the temperature, the lower the thermal noise. Nevertheless, a change of 20-30°C in temperature doesn't make any difference on the decibel scale.

The natural thermal noise power of a GSM radio channel is of -120 dBm. However, this power mostly depends on the channel's bandwidth.

Note: A difference between the natural thermal noise power of a GSM radio channel of -120dBm and the sensitivity of a BTS of -106 dBm can be explained in two ways:
 * the unavoidable differences in the receiver
 * the signal to noise ratio requires for the signal to be decoded correctly

Frequency Stability
Frequency stability represents, as its name suggests, the stability in frequency over time, or the measure to which the signal can produce a stable frequency for an certain amount of time.

Frequency Drift
Frequency drift is a non-linear phenomenon that causes unwanted progressive changes in the frequency in time.

Frequency drift can cause interference when, for example, a radio station switches to an adjacent channel. This phenomenon appears when the radio components are old or have flaws or during thermal changes.

Phase Instability
Phase instability, or phase noise, refers to the speedy, short-term variations caused by a signal's random frequency variations. In a spectrum analyzer, these fluctuations would appear as a noise spectrum on either side of the signal.

It is common phenomenon and it is widely believed that all signals experience a certain amount of phase noise. When the phase noise is too strong, it affects the signal quality and may increase the bit error rate in radio systems using phase modulation.