Insights from industry

Understanding the Telecom Filters of Today

insights from industryXiaolun ZengProduct Group ManagerIridian Spectral Technologies

In this interview, AZoM talks to Xiaolun Zeng, Product Manager of Telecoms at Iridian Spectral Technologies, about telecom filters, why they're important and how the field has advanced in recent times.

Please could you start by giving our readers a brief introduction to Iridian and the company’s work?

Iridian is a supplier of custom optical filters and coating solutions. The company was established in 1998 and is based in Ottawa, Ontario, Canada. It employs about 160 staff with extensive technical expertise in all areas of thin-film manufacturing. Our flagship facility was finished in November 2012 and is about 45,000 square feet in size.

Iridian achieved the ISO9001-2015 certification in May 2016, and we are also registered with the Canada Controlled Goods Program.

We design and manufacture thin-film dielectric filters and coatings, and we can offer these with a wavelength range from 300 nm to 15 µm. As well as customized solutions, we can provide single, multi-band and multi-zone filters with dimensions ranging from less than one square millimeter to more than 150 millimeters in diameter.

Much of our technology is based on energetic sputtering and evaporation processes. We have more than 20 coating machines available and custom-designed control software and in-house photolithography capabilities. 

Iridian - Today's Telecom Filters

Iridian - Today's Telecom Filters from AZoNetwork on Vimeo.


How has Iridian’s work in the telecom filters market developed over time, and where are these filters typically employed?

Iridian is a world leader in providing telecom filters for the long haul, X-haul, metro and access market. These markets tend to feature WDM in telecom channels, dual-band filters for wireless backbone systems, gain flattening filters and more.

Our filters are typically involved in the whole optical fiber optical communication system, from the light source and multiplexing devices to the fiber. The fiber often needs gain amplification, and the gain amplification device needs gain flattening filters. On the receiving side, the de-multiplexing device also needs our filters to separate the channels through the detection process.

Where do WDM channel skip filters fit into the wider telecom filter space?

WDM channel skip filters are one of our most popular products. These components are added to wavelength division multiplexing systems; for example, a WDM add/drop module in CWDM and DWDM applications or a module designed to facilitate band splitting and manage multiple ITU channels.

These filters feature very narrow transitions from the pass band to the blocking band to minimize lost channels while maintaining high spectral efficiency. This is ensured because the express channels only undergo one reflection.

What are the key considerations when designing and implementing WDM channel skip filters?

The designs of WDM channel skip filters are relatively simple. We have a CWDM skip filter and the DWDM skip filter, each with its own considerations.

The CWDM skip filter tends to only have a 2skip0 and 4skip0 configuration, while the DWDM skip filter offers more choice of filter type. Various wavelengths are available for a multiplexing signal, and we can choose between two and sixteen channels as required. The skip channel will depend on the application and the design requirements - it could be zero, one or two.

It is important to select skip filters on the basis of a cost-performance analysis. Telecom infrastructure budgets are typically strict, and network requirements are generally inflexible, so designers must balance the cost versus the performance when integrating channel skip filters.

The first consideration is the number of channel “X” and how many channels must be accommodated. For example, designers may need to make a choice between using 8skip0 filters or 4skip0 filters. The 8skip0 filters may have a higher cost than a 4skip0 filter, but fewer 8skip0 filters would be required for a specific application.

Considerations around performance often center around the number of channels allowed per filter while maintaining the modules’ necessary performance. For example, eight channels may accumulate an acceptable insertion loss, while four channels may accumulate an acceptable insertion loss.

Under these circumstances, a designer may consider the skip1 filter, which would lose one channel from the nine channels - about 11% of the network capacity across those channels.

The trade-off in this application is lower component cost because the skip1 filter costs much less than skip0 filters. The skip0 filter allows us to regain those lost channels, but this will cost more.

This decision could be critical for long-haul applications. For example, high-efficiency carriers are important in marine cables installed at the bottom of the ocean or cables installed underground where it is impractical to access the cable after installation.

For local applications, however, it might be more cost-effective to simply add more fibers to the cable when needed.

How does Iridian help its customers to navigate these decisions around performance and cost?

Nearly every application requires customization to ensure an optimal price-performance balance.

For example, two customers may both require 8skip0 filters that cover the same channels, but these customers will likely have different needs in terms of the wavelength range being blocked, pass bandwidth, reflect bandwidth, transmissions, reflection and isolation.

Here, a standard off-the-shelf stock filter would not meet the customer’s needs, so when designing a custom channel skip filter, we leverage our extensive experience in meeting the needs of countless customers to offer customized solutions where possible.

What are some of the benefits and applications of multi-band filters?

In current WDM and PON module designs, single band filters and multi-band filters are used for the same purpose - only to permit a narrow wavelength range to pass through the filter and reject any wavelengths outside of that range.

Multi-band filters are used to transmit two or more bands. These bands may be comprised of a wide bandwidth, a narrow bandwidth or the standard CWDM channel bandwidth. Multi-band filters can be used to replace two or more single pass band filters, allowing us to build in a single component.

We can’t just fill the available space - we also need to optimize this. Iridian recognizes the market’s need for components designed to enable and improve the long-term evolution of data communication.

The market need for multi-band filters is driven by the technological demands of 5G wireless transmissions - from the middle wave band to the mm wave band. Consumer devices are shrinking while simultaneously increasing in terms of connectivity. We are seeing this in numerous technologies employed in military, consumer and medical devices.

The most popular application for multi-band filters is currently wireless station interconnection for 4G, LTE and 5G systems. Multi-band filters can be used in any CWDM system where there is a need to minimize the component count and footprint.

For example, this could be achieved using the GPON, XG-PON and XGS-PON filters, which leverage three specific wavelength channels - 1310, 1490 and the 1550 nanometers, respectively.

Multi-band filters are able to produce a flat pass band and maintain good reflection isolation in almost all use cases, but channel spacing must remain reasonable, and the dead bands must not be too narrow

For example, 20 nanometers is an achievable band space between pass band and pass band, but 50 nanometers or 60 nanometers between each adjacent pass band channel is preferable. If the two pass bands are too close to each other, it will make the filter behave more like a narrow notch filter.

Assuming the specification requirements are similar, the total cost for single band filters and the dual band filters is comparable. Multi-band filters tend to be slightly more expensive than single band filters, but this should be less than 20% higher than the single band channels while maintaining the same performance. That is the key advantage of multi-band filters.

The key advantage for the designer is a reduction in the overall footprint of the module housing. This is essential when these systems are used to install wireless interconnected stations for 5G networks, particularly due to the soaring cost of urban real estate.

System performance is improved using MBPFs. This reduces the number of components in the module and optimizes the use of space, dimensions and designs in terms of the number of fiber connections required to reduce insertion loss.

What kind of applications benefit from the use of dual pass filters?

Iridian’s experience in designing carriers extends to the design of dual pass filters, and the company’s diverse range of capabilities allows us to produce components to accommodate even the most difficult specifications. This is important because our customers tend to require a huge range of different specifications in their dual pass filters.

Channel spacing is also important in dual band filter design. For example, the NG-PON filter requires the two pass bands to be very close to one another. In this design, the central wavelength of these two pass bands tends to be 1545, 1600 or 1610 nm.

The XG-PON dual pass filters can accommodate very wide operating wavelength ranges with deep isolation in reflection bands between the two pass bands. These also offer very low ripples in the pass bands, ensuring very good performance.

While the NG-PON dual pass filter tends to have the two pass bands close to each other like a notch filter, its performance is still good. The isolation between the two pass bands cannot be too deep, however.

Specialized multi-band filter designs are also available. The 100G DWDM dual pass filter means that the overall design has some limitations, but this is able to pass two DWDM channels and reflect in the middle.

Iridian can accommodate very high volumes and supply a full range of different multi-band filters. We design and manufacture these to ensure optimized price-performance balance, particularly in terms of our WDM and PON modules.

We can develop a customer-designed prototype and produce dual pass, triple pass and four pass filters while maintaining a consistent quantity and excellent performance.

Could you tell our readers about Iridian’s hybrid gain flattening filter?

A gain flattening filter is used to flatten or smooth out unequal signal intensity over a number of the specified channels in the C-band, L-band and U-band. This unequal signal intensity usually occurs after an amplification stage – with an amplifier such as an EFDA and/or Raman amplifier.

The hybrid gain flattening filter combines the functionality of both our WDM and gain flattening filters into one component.

GFFs and WDM filters work well with optical filter amplifiers. For example, the EDFA module used to use an erbium-doped fiber to eliminate the need to transfer the signal from the optical to electrical components and back. We call this the ‘passive amplify.’

The optical signal is amplified within the fiber by the laser operating at 980 nanometer or 1480 nanometer wavelengths. This is used to excite the erbium atoms in order to emit a large number of the photons triggered by the much weaker incoming optical signal at the same wavelength as the incoming electrical signal.

Most telecom wavelengths operate in the C-band or the L-band. Sometimes this is also the pass band, meaning that the pump laser acting upon the signal will disrupt these precise wavelengths - a particular problem in fiber accepting multiple channels.

In the traditional EDFA module, a filter is used to flatten or even out the outgoing amplified signal. These filters operate in tandem with the 980 nm or 1480 nm wavelength blocking filter in order to minimize interference caused by the pump laser at these two wavelengths.

The hybrid GFF can combine the functionality of both the WDM and the GFF in one component, effectively blocking the pump laser at 980 nm and 1480 nm while still providing gain flattening for the optical signal amplification. Our hybrid GFF can also be designed to pass or block other wavelengths as required.

Neither the base GFF nor the WDM are trivial components, so combining these into a single component that costs less than two separate filters is highly advantageous.

As well as eliminating the cost of the separate filter, using the hybrid GFF system reduces system design complexity by cutting out a necessary component, helping to make systems more compact and allowing designers to accomplish more within the same space.

Iridian’s hybrid GFF performs very well in terms of quantifiable performance metrics - GFF performance is primarily defined by the specification of the peak-to-peak error function, and Iridian already has experience in the production of base GFFs with very good performance.

As with the base GFF, the deeper the modulation depths, the more difficult it becomes to meet the customer’s ideal peak-to-peak error function. There is, therefore, no compromise in performance when upgrading from a base GFF to a hybrid GFF.

How would you summarize Iridian’s approach and capabilities to potential new customers?

Iridian has a long history as the leading global supplier of optical filters for the telecommunication industry. We can provide competitive pricing from the prototype stage to high volume manufacturing, leveraging our world class design and manufacturing capabilities to achieve this.

We can consistently provide the best filter solution for customers’ package components and modules. Our DWDM channel skip filter, multi-band filter, hybrid GFF filter and other telecom filters are custom designed to meet all of today’s telecom system application requirements.

If your readers have any specific designs, ideas or concepts in mind, I would encourage them to contact us. We can find a solution for them in almost every case.

About Xiaolun Zeng

Xiaolun Zeng has a B.ENg in Automation Technology, an MSc. in Applied information and Computer Science, and more than 20 years’ experience assisting customers with a wide range of technology needs for telecommunications applications. Prior to joining Iridian in 2005, Xiaolun served as R&D and sales & marketing manager at Broadex; he also served in senior engineering, product development, and sales roles at AFOP, Santec, and JDS Uniphase. As Product Group Manager at Iridian, Xiaolun assists our telecom customers with optical filter solutions.

This information has been sourced, reviewed and adapted from materials provided by Iridian Spectral Technologies.

For more information on this source, please visit Iridian Spectral Technologies.

Disclaimer: The views expressed here are those of the interviewee and do not necessarily represent the views of Limited (T/A) AZoNetwork, the owner and operator of this website. This disclaimer forms part of the Terms and Conditions of use of this website.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Iridian Spectral Technologies. (2024, February 26). Understanding the Telecom Filters of Today. AZoOptics. Retrieved on July 14, 2024 from

  • MLA

    Iridian Spectral Technologies. "Understanding the Telecom Filters of Today". AZoOptics. 14 July 2024. <>.

  • Chicago

    Iridian Spectral Technologies. "Understanding the Telecom Filters of Today". AZoOptics. (accessed July 14, 2024).

  • Harvard

    Iridian Spectral Technologies. 2024. Understanding the Telecom Filters of Today. AZoOptics, viewed 14 July 2024,

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback
Your comment type

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.