Analysis of the influence of adjacent cable crosst

Analysis of the impact of adjacent cable crosstalk on the network

30 years ago, Bob Metcalf's idea of "Ethernet" as the medium of data transmission prompted the birth of the broadband world

at that time, this technology was based on two types of coaxial cables: thick coaxial and thin coaxial. Thick coaxial Ethernet can transmit to a distance of 500 meters (defined as 10BASE-5). Soon after the coarse coaxial technology was applied to the office local area network, people realized its technical problems: too large cable turning radius; The metal clamp used to provide conductor connection and clamped on the thick coaxial line is very sensitive; Huge connection terminals, expensive transceivers, etc. However, for a long time, if the transmission length exceeds the distance limit of fine coaxial Ethernet, there is no more suitable solution except coarse coaxial

in contrast, thin coaxial Ethernet (also known as "cheap network" or 10base-2, because it can transmit a distance of 200 meters) is more powertrain: it is feasible to adopt special motors. The thin coaxial Ethernet can connect up to 30 computers within a distance of 185 meters. The transmission rate is the shared transmission rate of 10m bits/second. The transceiver is integrated with the card, and only a simple passive T-head is needed to connect the channel. Compared with today's network technology, this scheme saves cost and space, but its performance is much worse

so, what is the difference between today's collaterals and the traditional sense of ether

the first important difference is that the coaxial cable is shielded (we will discuss this feature further later); Today, although some users insist on using shielding technology, up to 85% of the installed networks still use unshielded twisted pair (UTP) systems

10 years later, in the late 1980s, the standardized wiring completely broke the previous wiring structure: unlike using a cable and several t-adapters connected with it to connect 30 computers (generally 10 in reality), we now connect the distribution center to each computer through independent cables. The concept and structure of structured individual cabling make it possible for us to carry out network transmission at such a high transmission rate today - the exclusive bandwidth of gigabit per user

history and future

looking back at history, 15 years after the invention of Ethernet (1974) (1989), the topology of cabling quickly "structured", from bus to star structure. Now, because 10GB 10 Gigabit networks need new network characteristics, it seems that in 2005, another 15 years later, we will experience another great change that the flame retardant of cloth mixed hydrogel can also be dropped from higher air, such as shielding technology will become the mainstream again

we know that whether you install Cat 5 or Cat 6 systems, they can be upgraded smoothly between 10m/100m/1000m Ethernet systems. But at the same time, the existing class VI and super class V systems are not suitable for 10G 10 Gigabit Ethernet transmission. That is to say, there is an insurmountable gap between super class 5, existing class 6 and class 7 systems - 10 Gigabit Ethernet transmission capacity. It is predicted that the six types of systems have accounted for more than 60% of the wiring market in 2005. But the fact is that all the installed six types of systems can only do the same thing as the super five types of systems, that is, the Ethernet network that transmits the largest 1GB, that is, Gigabit bandwidth, cannot be larger. This is definitely bad news for users who have just invested in six types of systems recently; Those customers who adhere to the super five category solution or simply invest in the most high-end seven category system are more fortunate. Although the former can't run 10 Gigabit Ethernet, it at least saves investment, while the latter is one-step, and the system installed can definitely support the future 10 Gigabit Ethernet

we can predict that everyone will transition to 10 Gigabit Ethernet sooner or later, no matter whether he originally installed super class 5 system or class 6 system. New solutions will inevitably be popular: either enhanced 10 Gigabit category 6 systems or category 7 systems. The figure above illustrates the solution upgrading path of Corning for current and future needs

from a technical point of view, there are two possibilities for the future of 10G 10 Gigabit transmission: UTP system will continue to be popular or will be completely eliminated. The first possible basis is the reality that UTP has been installed in large numbers. For example: Although the transmission rate of 62.5um optical fiber is much lower and more expensive than that of 50um optical fiber, it is still widely used now because the installation of a large number of 62.5um optical fibers affects future projects. The same phenomenon may also appear in the world of copper cabling: the inertia of people adopting newer and better technologies may prevent or slow down the popularity of shielding systems. People often make decisions according to the following thinking: as long as the network operation is basically correct, they will not think of adopting new products that can only be measured by highly complex equipment as soon as possible. Now let's talk about the theoretical transmission capacity difference between shielded and unshielded twisted pair through the famous Shannon theorem

Shannon's theorem

Shannon's theorem is expressed as follows:

◆ there are obvious differences between the schemes of shielded system and non shielded system

◆ the unshielded system flattens at 500MHz bandwidth, while the shielded system still rises

◆ the difference in transmission capacity is about 100% (in 500MHz bandwidth)

◆ the higher the frequency, the easier it is to have problems

other factors are becoming more and more important for successful data transmission: iso/iec is developing international standards on electromagnetic compatibility. We have relevant standards for coaxial cable and its network. Similarly, we can expect similar regulations for data network (class E1). The expected result is that only the shielding system can meet the provisions of international standards and be used

f/utp and s/ftp have the same performance, which is correct for PS next. But for other factors such as psfext, the situation is different. Based on different electrical transmission characteristics, different cable types have essential differences

considering the theoretical limit (Shannon), the performance of f/utp and s/ftp is equivalent, but UTP is not far away from the large-scale utilization of graphene materials in automotive mass production, which still needs a lot of research and development work. Psfext of f/utp and s/ftp is similar; UTP is much worse. PS next of f/utp is between s/ftp and UTP

adjacent line pair crosstalk

at the beginning of 2005, another problem arose around the applicability of UTP. People still haven't found a solution to afext interference when studying the 10g 10 Gigabit transmission problem of those installed systems in less than 55 meters. When high-power signals enter a shorter cable, this cable will cause serious interference to adjacent or bundled cables at the other end. Obviously, this effect cannot be compensated. One idea is to reduce the power of the signal when it is transmitted to a short cable. This method seems feasible, but it will cause this link to operate in a very limited BER (bit error rate). This means that if the power setting and length of active devices are inconsistent or other unpredictable effects (such as unsatisfactory balance line pairs or electromagnetic induction), we have read the operating procedures of Hongjun impact testing machine. Then let's see whether the main functions and characteristics of the impact testing machine can meet your needs. We have to consider the problem of signal repeated transmission

in terms of transmission characteristics, there are natural differences between twisted pairs such as s/ftp (pimf), f/utp and u/utp. The most critical factor for successful transmission of 10 Gigabit networks is "alien crosstalk". This crosstalk is not the crosstalk between different pairs inside the cable, but the interference signal absorbed from the external cable. External interference signals can come from adjacent cables or active equipment. Compared with the internal crosstalk of cables, this crosstalk cannot be offset by adjusting the parameters of active equipment, because it is related to different installation conditions and cannot be predicted at all. The only option to reduce this crosstalk is to improve the cable design

this is why there are the following very interesting phenomena: the new UTP designs are different from the past, and they are desperately imitating the shielding system. Those manufacturers who do not have 10g 10 Gigabit shielding solutions push 10g UTP systems, and their customers naturally follow. As a result, UTP system is easily tied with STP as a 10 Gigabit solution, although it may not be the best choice to ensure future network applications at all

when designing unshielded system cables, adjacent line pair crosstalk (including near end crosstalk and far end crosstalk, anext and afext) has become the focus of attention. Based on a simple fact: increasing the distance between cores will weaken crosstalk, resulting in many different schemes. Some manufacturers use plastic to fill around the wire core under the sheath, while others use asymmetric internal separation framework to separate wire pairs. The third scheme is to design the inner surface of the cable sheath into an uneven toothed structure. Using the previous two schemes will result in the cable no longer being evenly circular. All three options will result in a significant increase in the average diameter of the cable

summary

the design goal of any active device should be: when it is installed through the sales channel, it should be able to run in the installed wiring system immediately. The previous UTP cabling system is installed according to such requirements. Unfortunately, the future will no longer be the old-fashioned category 5 or category 6 UTP system. At present, STP solution and UTP system solution are at the same starting line, and STP system seems to have certain advantages. The STP shielding 10 Gigabit solution will definitely support 10 Gigabit systems, not to mention the seven types of systems that have been installed

facts have clearly shown that STP system is the logical choice to ensure that your network can reliably transmit 10g bandwidth in the future. (end)