Understanding Mediumband: The Analogy Between "Wireless Channels" and "Metals".

1. Introduction

Wireless communication is the process of transmitting information from a wireless transmitter (TX) to a distant wireless receiver (RX). Strangely, wireless communication is, not only a business between TX and RX, but also inextricably intertwined with the propagation environment, which is the entity between the TX and RX [1].

The wireless communication would be reliable and can be made to be more efficient in terms of energy and spectral usage if the propagation environment is favourable for wireless communication. If it is unfavorable, on the other hand, the communication would be affected adversely. If the propagation environment is extremely unfavorable, which often happens, the communication would completely break down even with the state-of-the-art TXs and RXs.

Wireless channel is a more technical word often used in wireless communication literature, but simply refers to the propagation environment.

Fig. 1: A typical wireless transmitter and a wireless receiver, where s(t) and r(t) are the transmit and receive signals respectively.

Due to the distortions caused by the propagation environment and noise, r(t) is not equal to s(t). Furthermore, the distortion caused by the propagation environment on s(t) is random and could change rapidly [2]. Finding s(t) from r(t) as efficiently as possible is the main goal in wireless communication.

2. Analogy between "wireless channels" and "metals"

Metal exhibits magnetic attraction below its Curie temperature. For iron, it is 770°C. When iron's temperature rises above its Curie point of 770°C, it abruptly loses its magnetic attraction. The Curie temperature is named after French physicist and chemist Pierre Curie, who showed that magnetism is lost at a critical temperature. Something similar happens to the wireless channels as well [3].

When the signal (i.e. s(t)) bandwidth is low, otherwise when s(t) is narrowband, the random distortion caused by the wireless environment on the data signal s(t) is "unimodally distributed with zero mean" meaning the probability density function (PDF) of the random distortion has a peak at zero. See Fig. 2 for a depiction. This peak means the distortion caused by the propagation environment on s(t) is highly likely to be adverse for the overall quality of wireless communication.

Why is the peak in the PDF unfavorable for wireless communication? This peak signifies the fact that the probability of the distortion being very small is very high. However, the distortion affects s(t) as a multiplicative factor, which in turn make the detection of s(t) from r(t) is extremely difficult. In contrast to the literal meaning of "distortion", in wireless communication however, it is more favorable to have this distortion as large as possible, and thus the term "channel gain or channel coefficient" to denote this distortion. If this distortion is extremely small, the wireless communication is said to be in deep fade, which is highly unfavorable, and can even cause the wireless communication to completely break down.

Fig. 2: PDF of the random distortion caused by the propagation environment on s(t) has a peak at zero when the bandwidth of s(t) is low. The impulse response in the middle denotes the propagation environment.

When the signal bandwidth increases above a certain value, otherwise when s(t) is mediumband, the random distortion caused by the wireless environment on the data signal s(t) would abruptly be "bimodally distributed with zero mean" meaning the PDF of the random distortion has now a trench/narrow dip at zero. See Fig. 3 for a depiction. This trench means the distortion caused by the propagation environment on s(t) is significantly less likely to be adverse for the overall quality of wireless communication. It is because the distortion (or the channel gain/channel coefficient) is significantly less likely to be small, and in turn the wireless communication is significantly less likely to be in deep fading.

Fig. 3: PDF of the random distortion caused by the propagation environment on s(t) has a trench at zero when the bandwidth of s(t) rises above a certain threshold. The impulse response in the middle denotes the propagation environment.

As shown in Table 1, we can summarize the analogous properties of metals and wireless channels, where the temperature of the metal is analogous to the bandwidth of s(t), the magnetism of metal is analogous to the deep fading in the wireless channel, and Curie temperature is analogous to the boundary between the narrowband and mediumband on, what we call, the Delay Spread-Symbol Period plane. See Table 1 for more information.

Table 1: Analogous properties of metals (iron) and wireless channels.

3. Why does the distortion behave differently in different circumstances?

The answer requires some mathematical development in order to be complete. In simple terms however, in the narrowband case, the multipath components add freely. As a result of central limit theorem, the sum of multipath components, which quantitatively is the distortion, give rise to a Normal PDF with a peak at zero. Whereas in the mediumband case, the multipath components cannot add freely, and are randomly scaled. It is the addition of multipath components with random scaling that causes the trench at zero as shown in Fig. 3 [3].

4. Application of mediumband

In modern wireless communication systems like 4G, 5G, WiFi, the information bearing signal is a broadband waveform. That means, in simple terms, every time we send a message or make a call to someone we launch a broadband waveform into the air.

Using IFFT and FFT blocks at TX and RX respectively, these broadband waveforms are traditionally designed as a set of orthogonal narrowband waveforms. However, narrowband waveforms have many drawbacks. Because, their signalling rate is low, and as shown in Fig. 2, are exposed to highly unfavorable distortion.

As we saw in Fig. 3, it is more favourable to make broadband waveforms, not as a set of orthogonal narrowband waveforms, but as a set of “orthogonal mediumband waveforms”. It is because, mediumband waveforms do signalling at a faster rate, and are exposed to favorable distortion as shown in Fig. 3.

REFERENCES

1. ​D. A. Basnayaka, "Communicating in the mediumband: What it is and why it matters," IEEE Communication Magazine, Nov. 2024. (Open Access) 

   
   

2. "Wireless communication research: What is it exactly?," Online: https://www.basnayaka.org, May. 2025.

   
   

3. D. A. Basnayaka, "Introduction to mediumband wireless communication," in IEEE Open Journal of the Communications Society, vol. 4, pp. 1247-1262, May. 2023. (Open Access)